although if you are using XMLDOM with JavaScript you can code something like var n1 = uXmlDoc.selectSingleNode("//bookstore/book[1]/title/@lang"); and n1.text will give you the value "eng"

@KorayTugay, No, the first expression selects, doesn't "return" -- a set of nodes, and this set of nodes is not a string. A node is not a string -- a node is a node in a tree. An XML document is a tree of nodes. lang="eng" is just one of many textual representations of an attribute node that has a name "lang", doesn't belong to a namespace, and has a string value the string "eng" – Dimitre Novatchev Oct 22 '14 at

//Parent[@id='1']/Children/child/@name will only output the name attribute of the 4 child nodes belonging to the Parent specified by its predicate [@id=1]. You'll then need to change the predicate to [@id=2] to get the set of child nodes for the next Parent. However, if you ignore the Parent node altogether and use: //child/@name you can select name attribute of all child nodes in one go.

//Parent[@id='1']/Children/child/@name Your original child[@name] means an element child which has an attribute name. You want child/@name.

So far I have this XPath string: //Parent[@id='1']/Children/child[@name]

The document.evaluate() method evaluates an XPATH query/expression. So you can pass XPATH expressions there, traverse into the HTML document and locate the desired element.

Case in point: take this css selector: h1.header > a[rel~="author"] Its shortest functional XPath equivalent would be //h1[contains(" "+normalize-space(@class)+" "," header ")]/a[contains(" "+normalize-space(@rel)+" "," author ")] …which is both much harder to read and write. If you wrote this XPath instead: //h1[@class="header"]/a[@rel="author"] …you would incorrectly have missed markup like <h1 class="article header"><a rel="author external" href="/mike">...</a></h1>

CSS syntax is awesome for two reasons: It is an order of magnitude faster and less resource intensive than the more complex XPath. When what you want to find can be found with a css selector, a corresponding XPath query doing the same would most of the time be much longer and harder to read.

Interestingly, it uses multiple methods to locate the annotation within the cited page: absolute character count, text to search for, and XPath notation.